Coordinating Shared Tasks in Human-Robot Collaboration by Commands

Area of Science:

• Robotics
• Human-Computer Interaction
• Industrial Automation

Background:

• Human-robot collaboration is increasing in industrial settings.
• Current robot programming primarily focuses on automated motions, with limited support for interactive tasks and real-time coordination.
• Dynamic task requirements in collaborative environments necessitate flexible coordination mechanisms.

Purpose of the Study:

• To explore the use of commands for coordinating shared tasks between humans and robots in a collaborative workspace.
• To develop and evaluate both speech-based and graphical command interfaces for human-robot task coordination.

Main Methods:

• A knowledge-based architecture was developed, checking robot action suitability at runtime based on task and world state.
• Two interfaces were created: one utilizing speech commands and another using a graphical command interface.
• The system was tested on industrially relevant tasks including assembly, kitting, and hand-over operations.

Main Results:

• Both speech-based and graphical command interfaces effectively facilitated coordination in human-robot collaboration.
• The graphical interface proved advantageous in noisy industrial environments where speech recognition may be hindered.
• The knowledge-based approach ensured that robot actions were contextually appropriate for the shared task.

Conclusions:

• Command-based interaction, particularly with high-level commands, is an effective strategy for coordinating complex shared tasks between humans and robots.
• Graphical interfaces offer a robust alternative to speech for robot command and control in challenging industrial settings.
• Further development in interactive task coordination is crucial for advancing human-robot collaboration.